Projects / Programmes
Social Sciences Methodology, Statistics and Informatics
January 1, 2004
- December 31, 2008
| Code |
Science |
Field |
Subfield |
| 5.03.00 |
Social sciences |
Sociology |
|
| Code |
Science |
Field |
| S274 |
Social sciences |
Research methodology in science |
anketno zbiranje podatkov, kvaliteta merjenja, neodgovori, vzorčenje, informacijska družba, socialna omrežja, analiza velikih omrežij, bločni modeli, linearni statistični modeli, teorija odločanja, analiza svetovnega spleta, web mining
Researchers (14)
Organisations (1)
Abstract
Research group is working in the following researh areas:
1) MEASUREMENT PROBLEMS IN SOCIAL NETWORK ANALYSIS. Social network analysis has become one of the most promising approaches in numerous social fields (sociology, economy, communication studies, psychology, etc.). Main research objectives within this area the following research is planned:
a) The analysis of measurement quality (validity, reliability),
b) The survey mode effects (face-to-face, telephone, mail, Web),
c) The influence of name generators' order of precedence and limitation of the number of listed names,
d) The potentials of new technologies (Internet, mobile phone devices) for social networks measurements.
2) ALGORITHMS FOR NETWORK ANALYSIS. Research group is one of the leading groups worldwide in developing methods for analysis of large (over 100,000 vertices) networks, such as analysis of connections among members of informal organizations (criminal groups, business lobbies) and analysis of electronic and mobile communications in larger communities. Research has been conducted in the following key directions:
a) The development and implementation of algorithms for analysis of large networks,
b) The development and implementation of algorithms for network visualization,
c) The enlargement of interactive graphical presentation for unveiling and representing clusters of networks,
d) The blockmodeling for weighted and large networks,
e) The statistical analysis of large networks.
3) MODELING AND EVALUATION OF NONRESPONSE. One of the most serious problems in the study of contemporary survey methodology is the steady growth of nonresponse. Trends are not clear and linear, mostly because of the fact that low response rate does not automatically result in high nonresponse error. Research will be done in the following directions:
a) The study of trends and explaining of causes for nonresponse,
b) The development of models for predicting nonresponse error,
c) The impact of different methods for studying nonresposne on the sampling variance.
4) NEW TECHNOLOGIES IN SURVEY RESEARCH. In a few years, new technologies (Internet, mobile phone, interactive TV, palmtops, etc.) will become prevalent tools for data collection. However, their influence on quality and costs is relatively unclear. Research plans are:
a) The meta analysis and modeling of response rate in Web surveys,
b) The issues of reliability, validity and bias in Web-based surveys,
c) The modeling of costs and errors,
d) The potentials and limitations of surveying over mobile phones (interviewer- and self-administered).
5) APPLICATION OF MATHEMATICAL NODELS IN SOCIAL SCIENCES:
a) The applications of linear algebra in statistics,
b) The applications of utility theory in decision making,
c) The applications of non-additive measures and probabilities in statistics and social sciences.
6) METHODOLOGICAL PROBLEMS WITH MONITORING INFORMATION SOCIETY'S PHENOMENA.
Research is directed to the following objectives:
a) To determine optimal benchmarks (i.e. comparable indicators) and analysis of their measurement stability,
b) The methodological problems of monitoring information society (telework, e-commerce, e-learning, e-government, etc.) and problem of integral approach to usability evaluation of Web pages,
c) The measurement and conceptual problems with studying security, privacy and trust and the role of multivariate interactions within this field (e.g., digital divide).
7) MULTIMODE ANALYSIS OF DATA FROM WORLD WIDE WEB: World Wide Web contains variety of data, which require synthesis of three types of analyses: classical statistical analysis of individual cases, analysis of relational data (i.e. network analysis) and text mining. Main objectives:
a) The investigation of basic methodological principals in synthesis of different approaches,
b) The development of unified analytical tool for complete analysis of World Wide Web.
Significance for science
1. MEASUREMENT PROBLEMS IN NETWORK ANALYSIS pose an obstacle to the wider application of network analysis. Namely, the quality of measurement represents an important contribution to the growing number of applications of social network analyses within several fields. The results of this research help to understand these problems. They also contribute to outlining the solutions, which supports the expansion of network analysis in the social sciences. 2. SOCIAL NETWORK ANALYSIS has – due to stronger needs emerging within different scientific fields – been seeing exceptional growth in development around the world in the last few decades. Members of the programme group have been strongly involved in developing this field, contributing particularly to the areas of blockmodeling (generalised, weighted and multi-mode blockmodeling) and the development of social networks over time. The research results have been published in the most distinguished journals and publishers (Cambridge University Press) in this field. 3. THE ANALYSIS OF LARGE NETWORKS is expanding the existing potential of social network analysis with respect to the number of units included in the analysis. The developers of the PAJEK programme package have powerfully improved the capabilities of the programme regarding the analysis of very large networks. As a result, PAJEK is currently one of the leading global tools in this field. The authors have also improved the graphical presentation and interactive potential of the programme package. All of these improvements can significantly contribute to the global application of network analysis in numerous social scientific areas. 4. NONRESPONSE MODELING - research in this area contributes to solutions to one of the most pressing problems in contemporary survey research. The steady growth of the nonresponse rate has now become a serious threat to the quality of statistical inference. The models developed for analysing the factors influencing the level of nonresponse contribute to an understanding of the factors that determine nonresponse. They also offer a starting platform for prediction, prevention and post-survey adjustments. 5. NEW TECHNOLOGIES IN SURVEY RESEARCH is a topic that requires a permanent, ongoing scientific reflection. In particular, the evaluation of their potential is extremely important. The past methodological research in the area of Web surveys - performed in the projects RIS and WebSM - have already achieved a global response. The research work in this area brought new findings and further contributed to the increased use of new technologies in survey research. 6. THE APPLICATION OF MATHEMATICAL MODELS IN THE SOCIAL SCIENCES is a relatively new field of scientific research. Here, the methods developed for application in other fields were basically re-adapted for the social sciences. In addition, some new specific models and variations have been developed. The quality of the research findings and publications indicate that the group is also internationally recognised in this field. 7. METHODOLOGICAL ISSUES IN MONITORING THE INFORMATION SOCIETY: The research in this area is extremely important to allow proper international comparisons of information society indicators. Methodological problems frequently make these comparisons very difficult. The complexity of interactions among different variables may even lead to entirely wrong conclusions. The analysis contributed to a better understanding of these phenomena and to overcoming some of the challenges in cross-country comparisons. 8. MULTIMODE ANALYSIS OF THE WEB: The Internet and World Wide Web are becoming very important aspects of personal and business activities. This also poses new challenges for the social sciences. The research introduced a new analytical model for the simultaneous analysis of data from various sources (data mining, textual analysis, network analysis). Such an approach represents an entirely new way of addressing these issues.
Significance for the country
SUSTAINABLE SOCIO-ECONOMIC AND CULTURAL DEVELOPMENT 1) Impact on general quality of the survey instrument: Users in the public, non-profit and business sectors have an opportunity to use an improved instrument for survey measurements of different social indicators, from unemployment issues to political attitudes. The instrument is particularly useful for nonresponse problems and for questions related to social network analysis. 2) Impact on effectiveness of network measurements: The results are particularly useful for the analysis of large networks, including the analysis of relationships among companies/organisations, relations between subjects/persons and the relations among the users of modern technologies (the Internet, mobile phone devices, Web presentations). The analysis of links among informal organisations (e.g. criminal groups, business lobbies etc.) is also very important. 3) Impact on the introduction of new technologies in survey research: In developed countries use of new technologies in survey researches dramatically reduces the costs (by up to 70%). In some developed countries, the Web is already being used in around 20% of all surveys. The research is thus also highly relevant for Slovenia because it is relatively developed in the area of information technologies. 4) Impact on an improved understanding of the information society: The results have improved the general quality of indicators of the information society. This has also increased the level of the understanding of the related problems. Only a proper understanding of these issues enables the establishment of the optimal measures to regulate and stimulate developments in this area. 5) Impact on more effective analysis of information related to the Web: The results enable a quick and rich multimodal analysis of data related to specific websites, but also to the World Wide Web as a whole. With that, the new quality of more integrated business information is achieved. Such analyses can contribute to the more effective design of websites. TECHNOLOGICAL DEVELOPMENTS 1) The proposed research contributes to the higher quality and increased international comparability of the information society indicators already published on the RIS.org site. 2) The research is particularly relevant for a proper understanding of the factors determining information technology developments. The RIS project thereby continues to improve the quality of its research on information technology in Slovenia. The RIS reports are widely used by companies, governmental institutions and researchers. 3) The PAJEK programme for the analysis of large datasets also contributes to the general use of technological applications. The simultaneous multimodal analysis of the World Wide Web (MAW) is also on its way to becoming another information technology application enjoying global recognition. STRENGTHENING THE NATIONAL IDENTITY AND PRESERVING THE NATURAL AND CULTURAL HERITAGE 1) The PAJEK programme for large network analysis has become a globally accessible research product. The website with this programme is thus one of the few Slovenian websites which offers Slovenian research findings to the global scientific community. It thereby also contributes to the modern identity of Slovenia as a whole. 2) The English webpage of the RIS.org project offers detailed information on the situation regarding information technology in Slovenia. It accordingly promotes the whole country on the Web. 3) The web page WebSM, the leading global page for the field of Web survey methodology, was also designed in Slovenia, which offers some opportunity for the country’s corresponding global promotion.
Most important scientific results
Final report,
complete report on dLib.si
Most important socioeconomically and culturally relevant results
Final report,
complete report on dLib.si
